Phylogenetics1 PDF

Phylogenetics  What is phylogenetics? Study of branching patterns of descent among lineages Lineages – Populations – Species – Molecules Shift between population genetics and phylogenetics is often the species boundary – Distantly related populations also show patterning...

Phylogenetics  What is phylogenetics? Study of branching patterns of descent among lineages Lineages – Populations – Species – Molecules Shift between population genetics and phylogenetics is often the species boundary – Distantly related populations also show patterning – Patterning across geography  What is phylogenetics? Goal: Determine and describe the evolutionary relationships among lineages – Order of events – Timing of events Visualization: Phylogenetic trees – Graph – No cycles  Phylogenetic trees Nodes – Terminal – Internal – Degree Branches Topology  Phylogenetic trees Rooted or unrooted – Rooted: Precisely 1 internal node of degree 2 Node that represents the common ancestor of all taxa – Unrooted: All internal nodes with degree 3+ Stephan Steigele  Phylogenetic trees Rooted or unrooted – Rooted: Precisely 1 internal node of degree 2 Node that represents the common ancestor of all taxa – Unrooted: All internal nodes with degree 3+  Phylogenetic trees Rooted or unrooted – Rooted: Precisely 1 internal node of degree 2 Node that represents the common ancestor of all taxa – Unrooted: All internal nodes with degree 3+ Binary: all speciation events produce two lineages from one Cladogram: Topology only Phylogram: Topology with edge lengths representing time or distance Ultrametric: Rooted tree with time-based edge lengths (all leaves equidistant from root)  Phylogenetic trees Clade: Group of ancestral and descendant lineages Monophyly: All of the descendants of a unique common ancestor Polyphyly: Descendants include lineages from multiple ancestors Paraphyly: One or more monophyletic subgroups are left apart from all other descendants of a unique common ancestor Phylogenetic trees Phylogenetic trees Phylogenetic trees Phylogenetic trees Figure 17.16 The transition from genetic polyphyly to paraphyly to monophyly in speciation Conceptual framework Conceptual framework Conceptual framework Conceptual framework Conceptual framework  Conceptual framework In each lineage, new mutations are fixed independently Each subsequent mutation is placed on a the previously fixed sequence Given enough time, each lineage goes through a unique sequence of (nested) fixation events Sequential fixation generates a readable history of sequence similarity and differences How to read this sequence? – Given extant taxa, how to reconstruct history? – Use nested shared similarity to infer history  Building phylogenetic trees Distance: Estimate distance matrix given data, generate tree that represents these distances Parsimony: Attempts to find a tree that minimizes the number of changes given data Maximum likelihood & Bayesian: Model-based approach to find the most-likely tree given the data  Distance Compute a distance matrix given a set of biological data – UPGMA – Neighbor joining Compute a tree that most resembles this distance matrix  Distance Unweighted pair group method using arithmetic averages (UPGMA) Given a set of taxa and a distance matrix, UPGMA produces a rooted tree with edge lengths Clusters taxa, then merging clusters Assembled outside in Distance Distance  Distance Neighbor-Joining (NJ) Saitou and Nei 1987 Given distance matrix, produces an unrooted phylogenetic tree with edge lengths Repeatedly pairing neighboring taxa Determine which nodes are neighbors based only on distance matrix  Parsimony The preferred evolutionary tree is the one that requires “the minimum net amount of evolution” (Edwards and Cavalli-Sforza, 1963) Each taxon described by a set of characters Each character can be in one of a finite number of states Steps = changes in character states Goal: Find the tree that explains the distribution of character sets across taxa with the fewest number of steps Parsimony Teresa Przytycka  Parsimony 1 2 3 4 6 5 Teresa Przytycka  Parsimony Binary characters Multistate characters Ordered changes Reversible changes  Parsimony Fitch Wagner Dollo Carmin-Sokal prefect  Maximum likelihood Given a set of biological data, and a probabilistic model of evolution, find the tree that has the highest probability of generating the data – Multiple sequence alignment – Nucleotide substitution model  Tree Search methods Exhaustive search (exact) Branch and bound (exact) Heuristic (approximate)  Exhaustive Evaluate lengths of every possible tree Number of Taxa Number of trees 3 1 5 15 10 2,027,025 20 1020 50 1074  Branch and Bound Hendy and Penny 1982 Much faster, still guaranteed to find the best tree Determine upper bound of length of shortest tree Follow predictable search path through tree possible tree topology Abandon any fork of search tree where the upper bound is exceeded before the last taxon is added  Heuristic Search trees by swapping Very fast Find a starting tree – Stepwise addition – Star decomposition Rearrange tree to find better trees – Nearest neighbor Interchange (NNI) – Subtree pruning and regrafting (SPR) – Tree bisection and Reconnection (TBR)  Heuristic Bootstrap – Randomly resample the data with replacement – Rebuild tree – What fraction of the bootstrap samples show support for a particular node? Heuristic Felsenstein Efron et al. 1996  Heuristic Bootstrap – Randomly resample the data with replacement – Rebuild tree – What fraction of the bootstrap samples show support for a particular node? Jackknife – Randomly subset data – Rebuild tree – What fraction of jackknife samples show support for particular node – Whether excluding certain characters has major effect on tree  Populations of North American Bears  Software: PHYLIP Joe Felsenstein (1980) Over 29,000 registered users Parsimony, distance matrix, ML DNA, RNA, protein, restriction sites, discrete characters, continuous characters, allele frequencies, distance matrices Freely available http://evolution.gs.washington.edu/phylip.html Webservers available as well  Software: MEGA Molecular Evolutionary Genetic Analysis Sudhir Kumar Parsimony, distance matrix, ML – NJ – UPGMA – Minimum evolution Molecular data (nucleic acid, protein sequences) Bootstrapping, consensus trees Data editing Sequence alignment (with ClustalW) http://www.megasoftware.net  Software: EMBOSS European Molecular Biology Open Software Suite Peter Rice, Alan Bleasby, Jon Ison General sequence analysis Phylogeny (PHYLIP) Alignment (CLUSTAL) All sequence formats Many alignment formats http://emboss.sourceforge.net/what/